A Latent Analysis of Earth Surface Dynamic Evolution Using Change Map Time Series

Abstract

With a continuous increase in the number of Earth Observation satellites, leading to the development of satellite image time series (SITS), the number of algorithms for land cover analysis and monitoring has greatly expanded. This paper offers a new perspective in dynamic classification for SITS. Four similarity measures (correlation coefficient, Kullback-Leibler divergence, conditional information, and normalized compression distance) based on consecutive image pairs from the data are employed. These measures employ linear dependences, statistical measures, and spatial relationships to compute radiometric, spectral, and texture changes that offer a description for the multitemporal behavior of the SITS. During this process, the original SITS is converted to a change map time series (CMTS), which removes the static information from the data set. The CMTS is analyzed using a latent Dirichlet allocation (LDA) model capable of discovering classes with semantic meaning based on the latent information hidden in the scene. This statistical method was originally used for text classification, thus requiring a word, document, corpus analogy with the elements inside the image. The experimental results were computed using 11 Landsat images over the city of Bucharest and surrounding areas. The LDA model enables us to discover a wide range of scene evolution classes based on the various dynamic behaviors of the land cover. The results are compared with the Corinne Land Cover map. However, this is not a validation method but one that adds static knowledge about the general usage of the analyzed area. In order to help the interpretation of the results, we use several studies on forms of relief, weather forecast, and very high resolution images that can explain the wide range of structures responsible for influencing the dynamic inside the resolution cell.

Robust Change Captioning

“…Change detection One popular domain for image-based change detection is aerial imagery [35,54,63], where changes can be linked to disaster response scenarios (e.g. damage detection) [17] or monitoring of land cover dynamics [29,55]. Prior approaches often rely on unsupervised methods for change detection, e.g.…”

Joint SAR Image Time Series and PSInSAR Data Analytics: An LDA Based Approach

“…Furthermore, the probability of the corpus is measured as the product of the marginal probabilities of single documents: The obtained k-means classes will represent visual words. In order to avoid limitations in content representations (due to a high similarity threshold), tests and previous results recommends a number of 150-250 visual words [17]. At the next step, we associate a patch to the document.…”

“…One of the algorithms that proved to be very efficient for multispectral SITS analysis [17], but also for single scene classification, regardless the type of imagery [18][19][20], is the Latent Dirichlet Allocation (LDA) model. The secret lies in the way one defines analogies between the analyzed data and notions like word, document, and corpus.…”

Semantic analysis of EO-1 hyperion hyperspectral data

“…To apply LDA outside text domain the user must define a synthetic language, finding equivalents for words, documents and corpus, while respecting the bag-of-words assumption. In remote sensing analysis this is often achieved by extracting a set of features from data, such as mean and variance from patches of Quickbird data [16], of spectral signatures and fuzzy templates extracted from patches of Landsat and Quickbird data [17], or the analysis of satellite image time series by extracting change descriptors between pairs of images [18], followed by a vector quantization process, usually performed using k-means.…”